We have previously identified a region of linkage to both body mass index (LOD=3.6), age-adjusted diabetes (LOD=1.7) and the combined phenotype of "diabesity" (LOD=5.2)on chrom 11q23-24. We genotyped 10 markers within this region on 3,000 additional individuals who were selected for BMI, but were not necessarily from families informative for diabetes. There was no linkage to BMI in this second set of Pimas, indicating that the initial linkage was dependent upon an interaction between BMI and diabetes. All known candidate genes in the region from 11q22-24, including the Apo CIII/AIV/A2 gene cluster, two inwardly rectifying potassium channels, and the dopamine receptor DRD2 were analyzed by genotyping several single nucleotide polymorphisms (SNPs)in each gene. Several SNPs in DRD2 showed a significant association with BMI; therefore, this gene was sequenced in its entirety and two SNPs were genotyped in 1300 Pimas. Unfortunately, SNPs detected in DRD2 do not appear to be the underlying cause of the linkage to BMI observed in this region. We have also created a physical map by identifying overlapping YAC clones across the region of linkage on chrom 11q and have mapped all of our SNPs to this contig. We are currently contructing a much more precise BAC contig across this region. We have currently completed a BAC contig across the uppermost peak of linkage which spans 3 megabase of DNA. We are identifying SNPs at 25 kB intervals across this interval and genotyping them in 100 sib-pairs (200 individuals) who contributed the most to the linkage observed on chromosome 11. In addition SNPs are typed in 100 unrelated case/control samples for BMI and 100 unrelated case/control samples for diabetes. To date, we have identified and genotyped 106 SNPs. Three of these SNPs have shown significant associations (p<0.001) with all three phenotypes (diabetes, BMI and the combined "diabesity") and have consequently been genotyped in a larger set of 1300 Pimas. However, adjustment of the linkage peak for each of these SNPs indicates that none of these variants were the genetic basis for the original linkage finding. We are currently extending our SNP linkage disequilibrium mapping across the entire region of linkage.